Heavy and problematic alcohol use is widespread on U.S. college campuses. Self-medication theory (SMT) posits that individuals exposed to trauma and resulting posttraumatic stress symptoms (PTSD) are at risk for heavy drinking and associated negative consequences. College students are no exception. Estimates of PTSD among college students are comparable to rates in the general adult population, and recent research has linked PTSD and alcohol-related negative consequences in college individuals. Data also reveal that the social environment (e.g., perceived peer norms, active peer influences) is a critical predictor of drinking in college, and has been linked to heavy drinking and deleterious outcomes in students. Yet, research has also shown that emotionally supportive peers are critical to the diminution and resolution of PTSD symptoms. As such, the drinking behaviors of these individuals may be highly influential for college students struggling with posttraumatic stress, and the extent to which one's emotional support group may consist of peers who are heavily alcohol involved may have a negative effect on drinking outcomes. PTSD symptoms, alcohol consumption, and peer interactions also are dynamic, varying over time and in their association with one another. The present project aims to assess the dynamic relationship between PTSD, alcohol behavior, and the influence of emotionally supportive peers at the daily level. Specifically, this study will investigate: (1) whether higher daily levels of PSD symptoms are associated with increased levels of daily alcohol consumption and reported alcohol-related consequences within individuals; (2) whether daily PTSD symptom expression and alcohol behavior is moderated by the density of drinkers in the emotional support group; and (3) whether the relationship between daily PTSD symptoms and daily alcohol behavior will be moderated by the daily drinking behavior of people in the emotional support group. Trauma-exposed college students (N=100) reporting heavy drinking and various PTSD responses (no symptoms, moderate symptoms, high symptoms) will complete a baseline assessment of individual difference variables. Participants will then complete 30 days of daily web-based surveys assessing alcohol use and related consequences, PTSD symptoms, and emotionally supportive peer behavior (peer alcohol use, peer level of emotional support). Using event-level methodology in combination with hierarchical linear modeling techniques will allow for a fine- grained and ecologically valid prospective examination of study hypotheses. Moreover, assessing peer influences at the daily level among high-risk drinkers is novel. Results have implications for informing SMT processes as well as for college administrators aimed at intervening with high-risk problem drinkers.